Re: [Kawa-commonlisp-dev] [PATCH] Type refactoring.

[Charles Turner]

On 21 September 2012 21:32, Jamison Hope <address@hidden> wrote:
> Yeah, sorry about that. The day job was keeping me pretty busy for
> a while with low-level C and robots, which was making it hard to
> find time for the mental context switch to Kawa.

Mm, robots. Sounds like a lot of fun!

> 1. The only change to kawa/lang/Translator.java is to remove a
>    javadoc comment on checkDefaultBinding() which states
>    "The default implementation does nothing." But the implementation
>    of that method is still <code>{ return null; }</code>, so the
>    comment still seems to apply. I recommend taking Translator.java
>    out of the diff.

I originally wrote that comment, and decided is was superfluous, but
maybe I'm wrong.

> 2. It looks like the only changes to Language#getLangTypeFor(Type)
>    are indentation? Let's keep that as a separate change, too.
>    At some point when Per's got some free time (haha) maybe we can push
>    through a few untabify/indent patches to make Kawa source whitespace
>    more consistent.

This one's indentation is *really* screwed up, as in it's a very
noticeable formatting mistake. Not just GNU style vs. Java style.
Thought I'd throw it in the pot, I'll remove it tomorrow then. My
Emacs configuration changed to delete trailing whitespace on saves,
which caused some diff's I didn't want, might have missed a few lines,
but I think trailing whitespace is also not good, and unfortunately
Netbeans loves its trailing space, with no obvious way to make it
stop.

> 3. I'm not a huge fan of how CommonLisp's booleanType is instantiated
>    in both getTypeFor(Class) and getNamedType(String). It seems like
>    one of those methods should just defer to the other. Perhaps this
>    would work (haven't actually tried it):

Oops, I meant to clean that up before sending, thanks!

Charles.